JP2010134318A - Image capturing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for capturing an image whose blur is easily corrected. <P>SOLUTION: The image capturing method for capturing the image of a first part 9 by using an image capturing apparatus 32 includes: a first approaching process where the image capturing apparatus 32 approaches and stops at a scheduled clamping place 55 where the image capturing apparatus 32 images the first part 9; an image capturing process for capturing an image of the first part 9 by the image capturing apparatus 32, to form the image; and an image correcting process for correcting the image. In the first approaching process, the image capturing apparatus 32 is linearly moved with respect to the first part 9. Further, an attenuation standby time 65 being a time until the image capturing is started in the image capturing process, after the image capturing apparatus 32 approaches and stops at the scheduled clamping place 55 in the first approaching process is set to a time longer than an attenuation time when the vibration of the image capturing apparatus 32 is attenuated to a predetermined amplitude. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an imaging method, and particularly to a method for imaging an image that is easy to correct.

  In order to operate the workpiece, a method of capturing the workpiece and detecting the shape and position of the workpiece from the captured image is used. When the moving workpiece is imaged or when the imaging device moves, the imaging device may vibrate and blur may occur in the captured image. A method for correcting a blurred image is disclosed in Patent Document 1. According to this, the luminance profile of the image data is differentiated. Then, the blur start point and blur end point of the image are estimated. Thereafter, the blur of the image is removed using the blur start point and the blur end point.

JP 2003-198200 A

  After the relative position between the workpiece and the imaging device is moved in a plurality of directions, the workpiece and the imaging device are stopped at a position facing each other. At this time, the relative position between the workpiece and the imaging device vibrates in a plurality of directions. When an image is captured in this state, the captured image is blurred in a plurality of directions. It is difficult to form an image without blur using an image in which blur is formed in a plurality of directions. Therefore, a method of capturing an image that can easily correct the image has been desired.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following forms or application examples.

[Application Example 1]
The imaging method according to this application example is an imaging method for imaging a workpiece using an imaging device, wherein the imaging device approaches and stops at an imaging location where the imaging device images the workpiece, and the imaging device Has an imaging step of imaging the workpiece to form an image, and an image correction step of correcting the image, wherein the imaging device linearly moves with respect to the workpiece in the first approaching step. And

  According to this imaging method, the imaging device moves linearly and approaches the imaging location. The vibration component of the imaging device is attenuated while the imaging device moves to the imaging location. And when an imaging device stops, since an acceleration is added to an imaging device, an imaging device vibrates. At this time, since the imaging apparatus stops from the linear movement, the direction in which the imaging apparatus vibrates becomes larger in one direction in which the linear movement is performed. Therefore, the image to be captured is an image that is likely to be blurred in one direction. As a result, in the image correction process, it is possible to make correction easier than when correcting an image in which blurring occurs in multiple directions.

[Application Example 2]
The imaging method according to the application example includes a vibration state estimation step for estimating a vibration state in which the imaging device vibrates when the image is taken, and the image correction step corrects the image using the estimated vibration state. It is characterized by doing.

  According to this imaging method, since the image is corrected using the estimated vibration state information, it can be corrected with high quality.

[Application Example 3]
In the imaging method according to the application example described above, the imaging apparatus includes a second approach process in which the imaging apparatus moves to a linear movement start place that is a place where the linear movement is started, and the vibration state estimation process includes the linear movement start place The vibration state is estimated using information on a motion state of the imaging device.

  According to this imaging method, the vibration state is estimated from the motion state of the imaging device at the linear movement start location. Therefore, the vibration state estimation process can be performed before the imaging process. Since the vibration of the imaging device is attenuated even during the vibration state estimation step, an image with less blur can be captured.

[Application Example 4]
In the imaging method according to the application example described above, the motion state information is transition information of a moving speed of the imaging device.

  According to this imaging method, the vibration state is estimated using the transition of the speed when the imaging apparatus moves to the linear movement start location. Therefore, it is not necessary to use a device that detects vibration, speed, acceleration, and the like of the imaging device. Therefore, the blur can be corrected using a simple device.

[Application Example 5]
In the imaging method according to the application example, the motion state information is information obtained by detecting a speed or acceleration of the imaging device.

  According to this imaging method, since the speed or acceleration of the imaging device is detected, it is possible to accurately grasp the vibration state when the imaging device is located at the linear movement start location. Therefore, it is possible to accurately estimate the vibration state of the imaging apparatus in the imaging process.

[Application Example 6]
In the imaging method according to the application example, the attenuation waiting time, which is a time until the imaging starts after moving from the linear movement start location in the first approach step, is attenuated to a preset amplitude. An imaging method characterized in that the time is set to be longer than the decay time.

  According to this imaging method, the attenuation standby time is set longer than the attenuation time. Therefore, the amplitude of vibration of the imaging device at the linear movement start location can be made smaller than the preset amplitude in the imaging process.

[Application Example 7]
In the imaging method according to the application example, the imaging device linearly moves in the second approach step.

  According to this imaging method, the imaging device moves linearly and moves to a linear movement start location. Therefore, it is possible to reach the linear movement start place with a short movement distance.

Hereinafter, embodiments will be described with reference to the drawings. In addition, each member in each drawing is illustrated with a different scale for each member in order to make the size recognizable on each drawing.
(Embodiment)
A characteristic imaging method and a manufacturing method for picking and assembling an imaged work in the present embodiment will be described with reference to FIGS. Picking indicates an operation of moving a workpiece by gripping the workpiece, moving it, and releasing it.

  FIG. 1 is a schematic perspective view showing the configuration of the assembling apparatus. As shown in FIG. 1, the assembling apparatus 1 mainly includes a component supply device 2, a robot 3, an assembly table 4, and a material removal conveying device 5. The component supply device 2 mainly includes a first component supply device 6, a second component supply device 7, and a third component supply device 8. The assembling apparatus 1 also has a function of a picking apparatus. The first component supply device 6 is a device that supplies a first component 9 as a workpiece, and the second component supply device 7 is a device that supplies a second component 10 as a workpiece. And the 3rd component supply apparatus 8 is an apparatus which supplies the 3rd component 11 as a workpiece | work.

  The first component supply device 6 includes a first component alignment device 12 and a first transport device 13. The first component aligning device 12 includes a conical dish portion 12a and a support base 12b that supports the dish portion 12a. And the vibration apparatus which is not shown in figure is arrange | positioned between the plate part 12a and the support stand 12b. A spiral step is formed inside the dish portion 12a. The step has a flat portion having a predetermined width, and the flat portion is a passage through which the first component 9 passes. The flat part is continuously formed from the bottom to the top of the dish part 12a. When the vibration device vibrates the dish portion 12a, the first component 9 moves along the flat portion. Since the width of the flat portion is formed such that only one first part 9 can pass, the first parts 9 are arranged in a row when the first part 9 passes through the passage.

  A belt conveyor 13 a is disposed on the upper side of the first transport device 13. The belt conveyor 13a extends long in one direction. This direction is the Y direction. In the horizontal direction, the direction orthogonal to the Y direction is defined as the X direction, and the vertical direction is defined as the Z direction. The 1st conveyance apparatus 13 equips an inside with a step motor and a pulley, and can move and stop the belt conveyor 13a. One end of the belt conveyor 13 a is connected to the upper part of the first component aligning device 12. The 1st component 9 which moved to the upper part of the plate part 12a moves on the belt conveyor 13a. The first component 9 is sequentially moved to the right side in the figure by the belt conveyor 13a and stopped at a predetermined place. Accordingly, the first components 9 are arranged and arranged on the belt conveyor 13a.

  A second component supply device 7 is disposed on the lower left side of the first component supply device 6 in the drawing. The second component supply device 7 includes a second component alignment device 14 and a second transport device 15. The second component aligning device 14 is the same device as the first component aligning device 12, and the second transport device 15 is the same device as the first transport device 13. And the belt conveyor 15a is arrange | positioned above the 2nd conveying apparatus 15, and the 2nd components 10 are arranged and arrange | positioned on the belt conveyor 15a.

  A third component supply device 8 is arranged on the upper right side of the first component supply device 6 in the drawing. The third component supply device 8 includes a third component alignment device 16 and a third transport device 17. The third component aligning device 16 is the same device as the first component aligning device 12, and the third transport device 17 is the same device as the first transport device 13. And the belt conveyor 17a is arrange | positioned above the 3rd conveying apparatus 17, and the 3rd components 11 are arranged and arrange | positioned on the belt conveyor 17a.

  A robot 3 is arranged on the right side of the component supply device 2 in the drawing. The robot 3 includes a base 20, and a turntable 21 is disposed on the base 20. The turntable 21 includes a fixed stand 21a and a rotation shaft 21b. The turntable 21 includes a servo motor and a speed reduction mechanism inside, and can rotate and stop the rotation shaft 21b with high angular accuracy.

  A first joint 22 is disposed in connection with the rotation shaft 21 b of the turntable 21, and a first arm 23 is disposed in connection with the first joint 22. A second joint 24 is disposed in connection with the first arm 23, and a second arm 25 is disposed in connection with the second joint 24. The second arm 25 includes a fixed shaft 25 a and a rotation shaft 25 b, and the second arm 25 can rotate the rotation shaft 25 b about the longitudinal direction of the second arm 25. A third joint 26 is disposed in connection with the rotation shaft 25 b of the second arm 25, and a third arm 27 is disposed in connection with the third joint 26. The third arm 27 includes a fixed shaft 27a and a rotation shaft 27b, and the third arm 27 can rotate the rotation shaft 27b with the longitudinal direction of the third arm 27 as a rotation axis. A hand portion 28 is disposed in connection with the rotation shaft 27 b of the third arm 27, and a pair of finger portions 28 a are disposed on the hand portion 28. The hand portion 28 includes a servo motor and a linear motion mechanism driven by the servo motor. The distance between the finger portions 28a can be changed by this linear motion mechanism.

  A first support arm 29 is arranged in connection with the rotary shaft 25b. The first support arm 29 is disposed so as to protrude above the second arm 25. A support joint 30 is disposed in connection with the first support arm 29, and a second support arm 31 is disposed in connection with the support joint 30. An imaging device 32 is disposed on the second support arm 31.

  The first joint 22, the second joint 24, and the third joint 26 each include a servo motor and a speed reduction mechanism, and can rotate and stop the first arm 23, the second arm 25, and the third arm 27 with high angular accuracy. it can. As described above, the robot 3 includes many joints and a rotation mechanism. In addition to these joints and the rotation mechanism, it is possible to grip the workpiece by controlling the finger portion 28a.

  Similarly, the support joint 30 includes a servo motor and a speed reduction mechanism inside, and can rotate and stop the second support arm 31 with high angular accuracy. Then, by controlling the angle of the second support arm 31 corresponding to the angle of the second arm 25, the optical axis of the imaging device 32 can be in the Z direction.

  An assembly table 4 is arranged on the upper right side of the robot 3 in the figure. The upper surface of the assembly table 4 is a work surface 4a, which is formed horizontally. Then, on the work surface 4a, the robot 3 can assemble the combined product 33 by combining the first component 9, the second component 10, and the third component 11.

  A material removal conveying device 5 is arranged on the upper right side of the assembly table 4 in the drawing. A belt conveyor 5 a is disposed on the upper side of the material removal conveyance device 5. The material removal transport device 5 is the same device as the first transport device 13, and when the robot 3 places the combined product 33 on the belt conveyor 5 a, the combined product 33 is moved by the belt conveyor 5 a.

  A control device 34 is arranged on the lower left side of the robot 3 in the drawing. The control device 34 is a device that controls the assembly device 1 including the component supply device 2, the robot 3, the material removal conveyance device 5 and the like.

  FIG. 2 is an electric control block diagram of the assembling apparatus. In FIG. 2, a control device 34 as a control unit of the assembling apparatus 1 includes a CPU (arithmetic processing device) 37 that performs various arithmetic processes as a processor and a memory 38 as a storage unit that stores various information.

  The robot drive device 39, the imaging device 32, the first component supply device 6, the second component supply device 7, the third component supply device 8, and the material removal transport device 5 are connected to the CPU 37 via the input / output interface 40 and the data bus 41. It is connected to the. Further, the input device 42 and the display device 43 are also connected to the CPU 37 via the input / output interface 40 and the data bus 41.

  The robot drive device 39 is a device that is connected to the robot 3 and controls the operation of the robot 3. The robot drive device 39 can output information on the posture of the robot 3 to the CPU 37. Then, the imaging device 32 can be moved to a location designated by the CPU 37 and a desired location can be imaged. Furthermore, it is possible to hold the work by moving the hand part 28 to a place designated by the CPU 37 and operating the finger part 28a.

  The input device 42 is a device for inputting operation conditions such as an operation for placing a workpiece and an assembly operation. For example, it is a device that receives and inputs coordinates indicating the shape of each workpiece from an external device (not shown). The display device 43 is a device that displays data and work status related to the workpiece and the robot. An operator performs an input operation using the input device 42 based on the information displayed on the display device 43.

  The memory 38 is a concept including a semiconductor memory such as a RAM and a ROM, and an external storage device such as a hard disk and a DVD-ROM. Functionally, a memory area for storing the program software 44 in which the operation control procedure in the assembling apparatus 1 is described is set in the memory 38. Furthermore, a storage area for storing work-related data 45 that is information such as the shape of the first component 9, the second component 10, and the third component 11 and the location where the hand portion 28 is gripped is also set in the memory 38. Further, robot-related data 46 that is information such as the relative positions of the first component supply device 6, the second component supply device 7, the third component supply device 8, the assembly table 4, the material removal transfer device 5 and the robot 3. A storage area for storage is also set in the memory 38. In addition, a memory area for storing vibration amount data 47 that is information such as attenuation characteristics at which the vibration of the imaging device 32 attenuates after the robot 3 moves and stops the imaging device 32 is also set in the memory 38. Furthermore, a storage area for storing image filter data 48 used for correcting an image captured by the imaging device 32 is also set in the memory 38. In addition, a memory area that functions as a work area for the CPU 37, a temporary file, and other various memory areas are set in the memory 38.

  The CPU 37 performs control for moving the workpiece after detecting the position and posture of the workpiece in accordance with the program software 44 stored in the memory 38. As a specific function realization unit, a robot control unit 49 that performs control for driving the robot 3 to move the workpiece is provided. In addition, it has a vibration calculation unit 50 that calculates a transition in which the vibration of the imaging device 32 attenuates using the speed and acceleration information of the imaging device 32. Furthermore, it has a filter calculation unit 51 that selects an image filter according to the vibration amount of the imaging device 32. Furthermore, it has the workpiece | work position calculating part 53 which calculates the position of a workpiece | work using the imaged image. In addition, it has a discharged material control unit 54 for controlling the operation of the belt conveyors 5a, 13a, 15a, and 17a in cooperation with the operation of the robot 3.

(Imaging method and robot control method)
Next, an imaging method and a robot control method will be described with reference to FIGS. 3 to 10 through an operation of assembling the first component 9, the second component 10, and the third component 11 using the assembly apparatus 1 described above. FIG. 3 is a schematic diagram for explaining the combined product, and the combined product indicates an article assembled by the assembling apparatus 1. FIG. 4A is a flowchart showing a part assembling process. FIG. 4B is a flowchart showing in detail a process in which the robot grips the part in the part assembling process. 5-10 is a schematic diagram for demonstrating the working method of an assembly operation.

  FIG. 3A is a schematic perspective view showing a combined product. FIG. 3B is a schematic plan view showing the combined product, and FIG. 3C is a schematic side sectional view taken along the line A-A ′ of the combined product in FIG. As shown in FIG. 3, the combined product 33 has a first part 9. The first component 9 is formed in a substantially rectangular parallelepiped shape. And a pair of groove part 9a is formed in the side surface of a X direction. And a pair of groove part 9a is formed in the place which opposes.

  A second component 10 is disposed on the first component 9. The second component 10 is formed in a substantially cylindrical shape. And a pair of groove part 10a is formed in the side surface of X direction both sides. The pair of groove portions 10a are formed at opposite locations. The groove portion 9a of the first component 9 and the groove portion 10a of the second component 10 are connected to each other.

  A third component 11 is disposed on the second component 10. The third component 11 is formed in a substantially portal shape. Accordingly, when both ends of the third component 11 are the end portions 11a, the pair of end portions 11a are located at opposite positions. The third component 11 is formed from a plate material having elasticity. The distance between the opposing end portions 11a is shorter than the opposing groove portions 9a and 10a.

  When assembling the combined product 33, first, the second component 10 is stacked on the first component 9 so that the groove 9a and the groove 10a are aligned. And the edge part 11a of the 3rd component 11 is inserted into the groove part 9a and the groove part 10a from an upper direction (Z direction). At this time, the opposite end 11a presses the groove 9a and the groove 10a. The third component 11 is difficult to come off in the Z direction due to friction between the end portion 11a and the groove portion 9a and the groove portion 10a. Therefore, the first component 9 and the second component 10 are not separated.

  In the flowchart shown in FIG. 4A, step S1 corresponds to a first component gripping step. The control device drives the first component supply device to place the first component on the belt conveyor. Next, the control device detects the position of the first component and causes the hand to grip the first component. Step S <b> 2 corresponds to a first component moving step, and is a step of moving the first component from the first component supply device to the assembly table. Next, the process proceeds to step S3. Step S3 corresponds to a second component gripping step. The control device arranges the second component on the belt conveyor of the second component supply device. Next, the control device detects the position of the second component and causes the hand portion to hold the second component. Next, the process proceeds to step S4. Step S4 corresponds to a second component assembly process. The control device moves the second component from the second component supply device to the assembly table. Next, the control device arranges the second component so as to overlap the first component. Next, the process proceeds to step S5.

  Step S5 corresponds to a third component gripping step. The control device arranges the third component on the belt conveyor of the third component supply device. Next, the control device detects the position of the third part and causes the hand part to hold the third part. Next, the process proceeds to step S6. Step S6 corresponds to a third part assembly process. The control device moves the third component from the third component supply device to the assembly table. Next, the control device inserts the third component from above the second component. Next, the process proceeds to step S7. Step S7 corresponds to an assembly moving process. The robot moves the combined product from the assembly table to the material removal conveyance device. Then, the material removal conveying device drives the belt conveyor to move the combined product. Step S8 corresponds to an end determination step and is a step of determining whether or not to end the assembly work. When the assembly work is continued, the process proceeds to step S1. When the assembling work is finished, the part assembling process is finished.

  FIG. 4B is a flowchart showing in detail the first component gripping step in step S1, the second component gripping step in step S3, and the third component gripping step in step S5. In the flowchart shown in FIG. 4B, step S11 corresponds to a second approach step, and is a step of moving the imaging device to a predetermined position. Next, the process proceeds to step S12. Step S12 corresponds to a vibration state estimation step, and is a step of estimating the vibration state of the imaging device when the imaging device images a component. Next, the process proceeds to step S13. Step S13 corresponds to a first approach step, and is a step in which the robot moves the imaging device to a place where the imaging device can image the component. Next, the process proceeds to step S14. Step S14 corresponds to an imaging step, and is a step in which the imaging device images a component. Next, the process proceeds to step S15. Step S15 corresponds to an image correction step, and is a step of correcting blurring of a captured image. Next, the process proceeds to step S16. Step S16 corresponds to a gripping process and is a process in which the hand grips the component. Thus, the component gripping process for each component is completed.

  Next, the imaging method and the robot control method in the assembly process will be described in detail with reference to FIGS. 5 to 10 in association with the steps shown in FIG. 5 to 9 are diagrams corresponding to the first component gripping step of step S1. Step S1 comprises steps S11 to S16, and will be described sequentially from step S11. FIG. 5A and FIG. 5B are diagrams corresponding to the second approach step of Step S11. As shown in FIG. 5A, in step S11, the first component 9 is placed on the belt conveyor 13a. Then, the belt conveyor 13a moves the first component 9. As a result, the first component 9 is positioned at a planned holding location 55 as a preset imaging location.

  Next, the robot 3 moves the imaging device 32 from the pre-movement location 56 to the imaging standby location 57 as a linear movement start location. The pre-movement place 56 is not a specific place, but is a place where the imaging device 32 is located when the work of the previous process is completed. The imaging standby location 57 is a preset location, and the distance 58 between the imaging standby location 57 and the planned holding location 55 is a preset distance 58.

  FIG. 5B is a time chart showing changes in speed and acceleration when the imaging device 32 moves from the pre-movement location 56 and stops at the imaging standby location 57. In FIG. 5B, the horizontal axis indicates the passage of time, and the time shifts from left to right. On the vertical axis, the speed and acceleration when the imaging device 32 moves are arranged. The speed and acceleration are larger on the upper side than on the lower side in the figure. A speed transition line 59 indicates a transition of speed until the imaging device 32 moves and stops. The acceleration transition line 60 indicates the transition of acceleration. As indicated by the speed transition line 59, the imaging device 32 moves at a constant speed for a predetermined time. This section on the time axis is defined as a constant speed section 59a. Next, a section where the speed transition line 59 decreases to 0 is defined as a deceleration section 59b.

  As the speed transition line 59 descends, the acceleration transition line 60 also descends. Then, as the speed transition line 59 approaches 0, the acceleration transition line 60 rises and approaches 0. The control device 34 controls the moving speed and acceleration of the imaging device 32 by controlling the motors arranged at each joint of the robot 3. Therefore, the speed transition line 59 and the acceleration transition line 60 can be controlled by the control device 34.

  FIG. 5C and FIG. 6A are diagrams corresponding to the vibration state estimation step of step S12. FIG. 5C shows how the imaging device 32 vibrates after the imaging device 32 stops at the imaging standby place 57. In FIG.5 (c), a horizontal axis shows progress of time and progress of time changes from the left to the right. The vertical axis indicates the displacement with which the imaging device 32 vibrates. A vibration transition line 61 indicates a line in which the transition of the displacement in which the imaging device 32 vibrates is predicted. The interval between the pair of envelopes 61a sandwiching the vibration transition line 61 is the amplitude of vibration. The vibration calculation unit 50 sets an amplitude threshold value 62 that is an amplitude threshold value in advance. The vibration calculation unit 50 calculates the vibration transition line 61 and the envelope 61a using information such as the mass distribution of the robot 3, the spring constant of each arm and joint of the robot 3, the speed transition line 59, and the acceleration transition line 60. . Next, an attenuation time 63 that is a time until the amplitude of the vibration transition line 61 becomes smaller than the amplitude threshold 62 is calculated.

  FIG. 6A illustrates a state in which the imaging device 32 vibrates after the imaging device 32 has moved to a location facing the planned holding location 55 and stopped. In FIG. 6A, the horizontal axis indicates the passage of time, and the passage of time changes from left to right. The vertical axis indicates the displacement with which the imaging device 32 vibrates. A vibration transition line 64 indicates a line in which the transition of the displacement in which the imaging device 32 vibrates is predicted. The movement start time 65a indicates the timing at which the imaging device 32 starts moving from the imaging standby location 57. At the movement start time 65a, an acceleration for moving the imaging device 32 is applied to the imaging device 32, so that the amplitude of the imaging device 32 increases. Since the imaging device 32 moves at a constant speed while moving, the amplitude of the imaging device 32 is attenuated. The timing at which the imaging device 32 stops is defined as a movement end time 65b. At the movement end time 65b, acceleration for stopping the imaging device 32 is applied, so that the amplitude of the imaging device 32 increases.

  The timing at which imaging is started after the imaging device 32 is stopped is defined as imaging start time 65c. When the time from the movement start time 65 a to the imaging start time 65 c is set as the attenuation standby time 65, the attenuation standby time 65 is set to be longer than the attenuation time 63. Therefore, it is difficult for the imaging device 32 to be affected by vibrations that occur when the imaging device 32 moves to the imaging standby place 57 and stops. The vibration calculation unit 50 calculates the amplitude 66 and the vibration direction at the imaging start time 65c. At this time, the vibration calculation unit 50 calculates the amplitude 66 and the vibration direction using conditions such as the vibration state of the image pickup device 32 in the image pickup standby place 57 and the moving speed and acceleration of the image pickup device 32 controlled by the control device 34.

  FIG. 6B is a diagram corresponding to the first approach step of step S13. As shown in FIG. 6B, in step S <b> 13, the robot 3 moves the imaging device 32 from the imaging standby location 57 to an imaging location that is opposite to the planned grip location 55. At this time, the control device 34 drives the robot 3 so that the imaging device 32 moves linearly. The control device 34 preferably drives the robot 3 so that the imaging device 32 does not vibrate left and right in the drawing. For example, it is preferable to drive the imaging device 32 by driving the first joint 22, the second joint 24, and the support joint 30 without operating the turntable 21. However, this method is not necessarily limited to when the influence of vibration is small.

  FIG. 6C is a diagram corresponding to the imaging process in step S14. In step S <b> 14, the imaging device 32 images the first component 9. As a result, as shown in FIG. 6C, an image 68 of the first component 9 is captured in the captured image 67. Since the image pickup device 32 picks up an image while vibrating in the Y direction, the image 68 is blurred in the Y direction. As a result, the side 68p on the Y direction side of the image 68 in the captured image 67 is observed to be thick.

  7 to 9 are diagrams corresponding to the image correction process in step S15. FIG. 7A shows an example of the correction filter 71. In step S15, blurring of the image 68 is corrected using a correction filter 71 as shown in FIG. The correction filter 71 is expressed in a matrix of 5 rows and 5 columns, for example. The number of rows and the number of columns are not limited to 5 rows and 5 columns, and may be set according to the degree of blurring.

  The rows of the correction filter 71 are the first row 71a to the fifth row 71e from the top in the figure. The columns of the correction filter 71 are defined as a first column 71f to a fifth column 71j from the left in the drawing. Each numerical value arranged in the correction filter 71 is referred to as an operator element 71k. The correction filter 71 is a filter called a sharpening filter. The operator element 71k in the third row 71c and the third column 71h located at the center of the matrix is set to a positive value. For example, in the embodiment, the value of the operator element 71k is set to 15. The operator element 71k around the third row 71c and the third column 71h is set to a negative value. For example, in the embodiment, the value of the operator element 71k is set to -1. The operator elements 71k of the first row 71a and the fifth row 71e in the third column 71h are set to a negative value larger than the other operator elements 71k. For example, in the embodiment, the value of the operator element 71k is set to -3. With this setting, strong sharpening can be performed in the Y direction. A value of 0 is set for the operator element 71k for which no numerical value is set in the operator element 71k of the correction filter 71. The value of the operator element 71k in the correction filter 71 is not limited to this.

  A plurality of correction filters 71 are stored in the image filter data 48 of the memory 38. Then, one correction filter 71 is selected in the vibration state estimation step of step S12. At this time, the number of rows and columns of the correction filter 71 and the value of the operator element 71k in each matrix of the correction filter 71 are set according to the predicted amplitude 66 of the imaging start time 65c. As for the relationship between the amplitude 66 and the pattern of the correction filter 71, an optimum combination is set in advance through experiments. If the setting of the correction filter 71 is not appropriate, it may be corrected to an image different from that of the first component 9, so that it is necessary to prepare an appropriate correction filter 71 by experimenting in advance.

  FIG. 7B shows an example in which the luminance of each pixel of the captured image 67 is displayed as a numerical value. The imaging device 32 includes an imaging element and an analog / digital conversion circuit. The imaging device 32 can output the luminance of each pixel of the image as numerical data of 256 gradations. In a place where light is not irradiated in the photographed image 67, the luminance value is output as 0. Then, the luminance value is output as 255 at the place irradiated with the brightest light. That is, in the captured image 67, bright pixels are large numerical values, and dark pixels are small numerical values. Each row of the image 68 is defined as a first row pixel 68a to a seventh row pixel 68g from the top to the bottom in the drawing. Then, each column of the image 68 is defined as a first column pixel 68h to a fifth column 68m from the left to the right in the drawing. Each numerical value arranged in the image 68 is referred to as a pixel element 68n.

  Next, an example in which the captured image 67 is corrected using the captured image 67 and the correction filter 71 will be described. In this example, the values calculated using the pixel elements 68n of the first row pixel 68a to the fifth row pixel 68e and the first column pixel 68h to the fifth column 68m are the pixels of the third row pixel 68c and the third column pixel 68j. The original 68n correction value is used. That is, the correction value of the pixel element 68n is calculated using the pixel elements 68n around the pixel element 68n to be corrected.

  First, the first row 71a of the correction filter 71 and the first row pixel 68a of the image 68 are used. The first column 71f and the first column pixel 68h are integrated, and the second column 71g and the second column pixel 68i are integrated. Similarly, the third column 71h and the third column pixel 68j are integrated, and the fourth column 71i and the fourth column pixel 68k are integrated. Similarly, the fifth column 71j and the fifth column 68m are integrated. Next, the sum total of the values integrated in each column is calculated. In the numerical values in the figure, 0 × 30 + 0 × 30−3 × 50 + 0 × 50 + 0 × 80 = −150.

  A similar calculation is performed using the second row 71b of the correction filter 71 and the second row pixel 68b of the image 68. In the numerical values in the figure, 0 × 80-1 × 50-1 × 70-1 × 70 + 0 × 80 = −190. Further, the same calculation is performed using the third row 71 c of the correction filter 71 and the third row pixel 68 c of the image 68. In the numerical values in the figure, 0 × 100-1 × 100 + 15 × 110-1 × 100 + 0 × 100 = 1450. Further, the same calculation is performed using the fourth row 71d of the correction filter 71 and the fourth row pixel 68d of the image 68. In the numerical values in the figure, 0 × 150-1 × 140-1 × 160-1 × 150 + 0 × 150 = −450. Further, the same calculation is performed using the fifth row 71e of the correction filter 71 and the fifth row pixel 68e of the image 68. In the numerical values in the figure, 0 × 180 + 0 × 170−3 × 190 + 0 × 180 + 0 × 190 = −570. Next, the sum of the values added in each row is calculated. In the numerical value in the figure, −150−190 + 1450−450−570 = 90. This value is the sum of the integrated values of 5 rows and 5 columns.

  Next, the sum of the operator elements 71k of the correction filter 71 is calculated. In the numerical values in the figure, the first line is 0 + 0-3 + 0 + 0 = -3. The second line is 0-1-1-1 + 0 = -3. The third line is 0-1 + 15-1 + 0 = 13. The fourth line is 0-1-1-1 + 0 = -3. The fifth line is 0 + 0-3 + 0 + 0 = -3. Next, the sum of the values added in each row is calculated. The sum of each row is −3-3 + 13−3-3 = 1. Next, the sum of the integrated values of the 5 rows and 5 columns is divided by the sum of the operator elements 71 k of the correction filter 71. In the numerical values in the figure, 90 ÷ 1 = 90. This value is set as a pixel conversion value. This pixel conversion value is used as the correction value for the pixel element 68n of the third row pixel 68c and the third column pixel 68j. The filter calculation unit 51 calculates the correction value of each pixel element 68n by performing the above calculation on each pixel element 68n.

  8 to 9 are graphs for explaining the operation of the correction filter 71, and FIG. 8A is a graph for explaining blurring of an image to be captured. In FIG. 8A, the horizontal axis indicates the position in the Y direction in the array of image sensors. The vertical axis represents the luminance distribution of the light irradiated on the imaging device 32. The brightness is higher on the upper side than on the lower side in the figure. The first luminance distribution line 72 indicates the luminance distribution at the start of imaging. A first luminance distribution line 72 indicates the distribution of reflected light from the first component 9. The first luminance distribution line 72 changes in distribution at the first position 72a, indicating that the first position 72a has a boundary between a bright place and a dark place. The first position 72 a corresponds to the location of the contour of the first component 9.

  A second luminance distribution line 73 indicates the luminance distribution when the imaging is finished. Similar to the first luminance distribution line 72, the second luminance distribution line 73 indicates the distribution of reflected light from the first component 9. The second luminance distribution line 73 changes in distribution at the second position 73a, indicating that the first position 72a has a boundary between a bright place and a dark place. The second position 73a corresponds to the location of the contour of the first component 9. That is, it shows that the outline of the first component 9 has moved from the first position 72a to the second position 73a between the start and end of imaging.

  The output of the imaging device 32 is an integrated value of the amount of light emitted to the imaging device during the imaging time. A third luminance distribution line 74 indicates the luminance distribution in the captured image 67 output from the imaging device 32. In the third luminance distribution line 74, the left side of the first position 72a has a small luminance because the amount of light irradiated is small. The right side of the second position 73a has a large luminance because the amount of light applied is large. The luminance increases from the left side to the right side between the first position 72a and the second position 73a. The change in luminance is gentle compared to the first luminance distribution line 72 and the second luminance distribution line 73, and is recognized as blurring. When the width between the first position 72a and the second position 73a is the pre-correction change width 74a, it is recognized that the longer the pre-correction change width 74a, the greater the blur.

  In FIG. 8B, the horizontal axis indicates the position in the Y direction in the numerical array of filters. The vertical axis indicates the value of the operator element. The value of the operator element is larger on the upper side than on the lower side. A filter outline 75 indicates a change in the operator element 71k in the third column 71h of the correction filter 71. The filter outline 75 has a positive peak value set in the third row 71c, and negative values set in the second row 71b and the fourth row 71d. Negative values are also set in the first row 71a and the fifth row 71e, and values lower than those in the second row 71b and the fourth row 71d are set. The filter outline 75 is set to be bilaterally symmetric and is set so as to be affected by the first row 71a and the third row 71c that are separated by two pixels with respect to the third row 71c.

  In FIG. 8C, the horizontal axis indicates the position in the Y direction in the array of image sensors. The vertical axis represents the luminance distribution after correcting the luminance of the light irradiated to the imaging device 32. The brightness is higher on the upper side than on the lower side in the figure. A corrected luminance distribution line 76 indicates a luminance distribution after the third luminance distribution line 74 is corrected using the correction filter 71 of the filter outline 75. The left side of the third position 76a in the post-correction divided luminance wiring 76 is flat. Since this place is not affected by the correction filter 71, the distribution is the same as the distribution of the third luminance distribution line 74.

  In the post-correction luminance wiring 76, the luminance at the location of the first position 72a is converted to be lower than the location of the third position 76a. In the third luminance distribution line 74, the luminance increases on the right side of the first position 72a. And since it is converted into a negative value by the first row 71a and the second row 71b of the correction filter 71, it is converted into a low value at the first position 72a.

  The right side of the fourth position 76b in the post-correction divided luminance wiring 76 is flat. Since this place is not affected by the correction filter 71, the distribution is the same as the distribution of the third luminance distribution line 74.

  In the post-correction luminance wiring 76, the luminance at the location of the second position 73a is converted to be higher than that of the location of the fourth position 76b. In the third luminance distribution line 74, the luminance decreases on the left side of the second position 73a. And since it is converted into a negative value by the fourth row 71d and the fifth row 71e of the correction filter 71, it is converted into a high value at the second position 73a.

  Conversion between the first position 72a and the second position 73a is performed so that the slope of the corrected luminance wiring 76 becomes steep. Therefore, conversion is performed between the first position 72a and the second position 73a so that the change in luminance becomes large.

FIG. 9A is a graph for explaining a method of further correcting the post-correction divided luminance wiring 76.
In FIG. 9A, the horizontal axis indicates the position in the Y direction in the arrangement of the imaging elements. The vertical axis represents the luminance distribution after correcting the luminance of light in imaging. The brightness is higher on the upper side than on the lower side in the figure. A post-correction divided luminance wiring 77 shows a luminance distribution after correcting a part of the corrected luminance wiring 76.

  Next, a method for correcting the corrected post-correction luminance wiring 76 and calculating the post-correction post-correction luminance wiring 77 will be described. The luminance on the left side from the third position 76a in the post-correction luminance wiring 76 is calculated, and the calculated value is set as the dark portion luminance value 76c. And the brightness | luminance of the place where the brightness | luminance lower than the dark part luminance value 76c in the post-correction part luminance wiring 76 is changed into the dark part luminance value 76c. Next, the luminance on the right side from the fourth position 76b in the post-correction luminance wiring 76 is calculated, and the calculated value is set as the bright portion luminance value 76d. Then, the luminance of the place having a higher luminance than the bright portion luminance value 76d in the post-correction luminance wiring 76 is changed to the bright portion luminance value 76d. As a result, the luminance in the vicinity of the first position 72a and the vicinity of the second position 73a is changed, and the corrected luminance wiring 76 is corrected to the recorrected luminance wiring 77. In the post-recorrection luminance wiring 77, the width in the Y direction where the luminance changes from the dark portion luminance value 76c to the bright portion luminance value 76d is defined as a post-correction change width 77a. Since the post-correction change width 77a is shorter than the pre-correction change width 74a of the third luminance distribution line 74, blurring can be reduced. Therefore, as shown in FIG. 9B, the side 79a on the Y direction side of the image 79 in the corrected image 78 is observed thinly. As a result, the workpiece position calculation unit 53 can detect the location of the side 79a in the Y direction with high positional accuracy.

  FIG. 10A is a diagram corresponding to the gripping step of step S16. As shown in FIG. 10A, in step S16, the hand portion 28 moves to the planned holding location 55. Then, the robot controller 49 holds the first component 9 by moving the finger part 28a. Since the workpiece position calculation unit 53 accurately detects the location of the first component 9, the finger portion 28 a can be held with the center of gravity of the first component 9 interposed therebetween. Therefore, the robot 3 can hold the first component 9 in a stable state. Step S1 is complete | finished above.

  FIG. 10B is a diagram corresponding to the first component moving step of step S2. As shown in FIG. 10B, in step S <b> 2, the robot control unit 49 drives the robot 3 to place the first component 9 on the work surface 4 a of the assembly table 4.

  In the second component gripping step in step S3, the second component 10 is gripped using the same method as in step S1. Therefore, since the workpiece position calculation unit 53 accurately detects the location of the second component 10, the finger portion 28 a can be held with the center of gravity of the second component 10 interposed therebetween. Therefore, the robot 3 can hold the second component 10 in a stable state.

  FIG. 10C is a diagram corresponding to the second component assembling process in step S4. As shown in FIG. 10C, in step S <b> 4, the robot control unit 49 drives the robot 3 to place the second component 10 on the first component 9. Since the hand portion 28 of the robot 3 holds the second component 10 with high positional accuracy, the robot 3 can place the second component 10 with high positional accuracy.

  In the third component gripping step in step S5, the third component 11 is gripped using the same method as in step S1. Therefore, since the workpiece position calculation unit 53 accurately detects the location of the third component 11, the finger portion 28a can be held with the center of gravity of the third component 11 in between. Therefore, the robot 3 can hold the third component 11 in a stable state.

  FIG. 10D is a diagram corresponding to the third part assembling process in step S6. As shown in FIG. 10D, in step S6, the robot control unit 49 drives the robot 3 to insert the third component 11 with the side surfaces of the second component 10 and the first component 9 interposed therebetween. Since the hand portion 28 of the robot 3 holds the third component 11 with high positional accuracy, the robot 3 can insert the third component 11 with high positional accuracy. As a result, the combined product 33 is completed.

  In the assembly moving process in step S7, the combined product 33 is gripped using the same method as in step S1. Then, the robot 3 moves the combined product 33 from the work surface 4 a onto the belt conveyor 5 a of the material removal transport device 5. Thereafter, the control device 34 moves the combined product 33 by driving the belt conveyor 5a. The part assembly process is completed by the above process.

As described above, this embodiment has the following effects.
(1) According to this embodiment, in the first approaching step of Step S13, the imaging device 32 moves linearly and approaches an imaging location that is a location facing the planned grip location 55. While the imaging device 32 moves to the imaging location, the vibration component of the imaging device 32 is attenuated. Then, since the acceleration is applied to the imaging device 32 when the imaging device 32 stops, the imaging device 32 vibrates. At this time, since the imaging device 32 stops from the linear movement, the direction in which the imaging device 32 vibrates is larger in one direction in which the linear movement is performed. Therefore, the image to be captured is an image that is likely to be blurred in one direction. As a result, in the image correction process in step S15, correction can be made easier than when correcting an image in which blurring occurs in multiple directions.

  (2) According to this embodiment, the attenuation standby time 65 is set longer than the attenuation time 63. Therefore, the influence of the amplitude of the imaging device 32 at the imaging standby place 57 can be made smaller than the amplitude threshold 62 in the imaging process of step S14.

  (3) According to the present embodiment, the pattern of the correction filter 71 is selected using the vibration state information estimated in the vibration state estimation step of step S12. And since the picked-up image 67 is correct | amended using the correction filter 71, it can correct | amend with sufficient quality.

  (4) According to the present embodiment, the vibration state is estimated from the speed and acceleration of the imaging device 32 in the vibration state estimation step of Step S12 at the imaging standby place 57. Therefore, the vibration state estimation process is performed before the imaging process of step S14. Since the vibration of the imaging device is also attenuated during the execution of the vibration state estimation step, it is possible to capture an image with less blur compared to the case where the vibration state estimation step is performed after the imaging step.

  (5) According to the present embodiment, the vibration state is estimated using the transition of speed or acceleration when the imaging device 32 moves to the imaging standby place 57. Therefore, it is not necessary to use a device that detects vibration, speed, acceleration, and the like of the imaging device 32. Therefore, vibration can be corrected with a simple apparatus configuration.

  (6) According to the present embodiment, the image is moved linearly from the pre-movement location 56 to the imaging standby location 57. Therefore, the imaging standby place 57 can be reached with a short moving distance.

In addition, this embodiment is not limited to embodiment mentioned above, A various change and improvement can also be added. A modification will be described below.
(Modification 1)
In the embodiment, the speed transition line 59 and the acceleration transition line 60 are calculated from the control data of the robot control unit 49, but the speed or acceleration may be detected. A speed or acceleration detection device may be arranged in the imaging device 32. The vibration state when the imaging device 32 is located at the imaging standby place 57 can be accurately grasped by the detection device. Accordingly, it is possible to accurately estimate the vibration state of the imaging device 32 in the imaging step of step S14.

(Modification 2)
In the said embodiment, the vibration transition line 61 and the envelope 61a are calculated in the vibration state estimation process of step S12. At this time, the vibration calculation unit 50 calculates the vibration transition line 61 and the envelope 61a using information such as the mass distribution of the robot 3, the spring constant of each arm and joint of the robot 3, the speed transition line 59, and the acceleration transition line 60. is doing. If the computation becomes complicated, experimental data may be used. The amplitude of the imaging device 32 is measured in advance by moving the imaging device 32 under various conditions. Then, the experimental data is stored. Then, the amplitude may be estimated from the speed data and acceleration data conditions.

(Modification 3)
In the embodiment, the captured image 67 is corrected using the correction filter 71. Other methods may be adopted as the correction method. For example, a general inverse filter function disclosed in JP-A-2006-279807 or a Wiener filter disclosed in JP-A-11-27574 may be employed. In addition, restoration methods such as a parametric Wiener filter, a restricted least square filter, and a projection filter disclosed in Japanese Patent Application Laid-Open No. 2007-183842 can be used. A correction method may be selected in accordance with the required accuracy.

(Modification 4)
In the above embodiment, a vertical articulated robot is adopted as the robot 3, but the robot is not limited to the form. Various types of robots such as a horizontal articulated robot, an orthogonal robot, and a parallel link robot can be employed.

The schematic perspective view which shows the structure of an assembly apparatus. The electric control block diagram of an assembly apparatus. (A) is a schematic perspective view which shows a united product, (b) is a schematic plan view which shows a united product, (c) is a schematic sectional side view which shows a united product. The flowchart which shows the assembly process of components. The schematic diagram for demonstrating the work method of an assembly work. The schematic diagram for demonstrating the work method of an assembly work. The schematic diagram for demonstrating the work method of an assembly work. The schematic diagram for demonstrating the work method of an assembly work. The schematic diagram for demonstrating the work method of an assembly work. The schematic diagram for demonstrating the work method of an assembly work.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 9 ... 1st part as a workpiece | work, 10 ... 2nd component as a workpiece | work, 11 ... 3rd component as a workpiece | work, 32 ... Imaging device, 55 ... Planned gripping place as an imaging location, 57 ... As a linear movement start location Imaging standby place, 68... Image.

Claims (7)

An imaging method for imaging a workpiece using an imaging device,
A first approach step in which the imaging device approaches and stops at an imaging location for imaging the workpiece;
An imaging step in which the imaging device images the workpiece to form an image;
An image correction step of correcting the image,
In the first approach step, the imaging device linearly moves with respect to the workpiece. The imaging method according to claim 1,
A vibration state estimation step of estimating a vibration state in which the imaging device vibrates when the imaging is performed;
An image pickup method comprising correcting the image using the estimated vibration state in the image correction step. The imaging method according to claim 2,
A second approach step in which the imaging device moves to a linear movement start location that is a location to start the linear movement;
The imaging method, wherein the vibration state estimation step estimates the vibration state using information on a motion state of the imaging device at the linear movement start location. The imaging method according to claim 3,
The imaging method characterized in that the information on the motion state is transition information of a moving speed of the imaging device. The imaging method according to claim 3,
The motion state information is information obtained by detecting a speed or acceleration of the imaging device. The imaging method according to claim 4 or 5, wherein:
In the first approaching step, the attenuation waiting time, which is the time until the imaging starts after moving from the linear movement start location, is set to be longer than the attenuation time in which the vibration of the imaging device attenuates to a preset amplitude. An imaging method characterized by comprising: The imaging method according to claim 6, wherein
In the second approaching step, the imaging device moves linearly.

Images